@Asmyldof mouser.com/Search/…

This is what you were telling me to look for, right?

yeah, that looks about right

Sheesh. I don't even know how to solder one of those bastards.

Looks like I'm going to get one shot per IC when prototyping too.

If I understand correctly...I should just be able to use the clock output on the same MCU I'm using to control this thing by serial, yes?

@Allenph if it's decently quiet, yeah. if its phase-noisy, then the PLL will just multiply up the phase noise :/

I have a bunch of PIC12F683s I was planning on using. I had assumed that if it was good enough for a processor it should be fine.

the clock oscillators on MCUs are probably not spec'ed for PN or jitter

Those pins are just for a reference frequency, right? There's a huge range of frequencies it says can be used as input. Crystals are cheap...think it's better to just throw one of those on the reference pins?

@Allenph if it has a built-in reference osc, that might be a better bet than what your uC has

@Shalvenay Maybe I just don't know enough to interpret the documentation then. I'm guessing uC means microcontroller.

@Allenph yeah

@Allenph I only said I saw plenty of stuff above the frequency you claimed was the limit. I still have no idea what you want it for, so I cannot even begin to tell you what to look for. But yeah, if you need stuff above a certain frequency it will be SMD with a middle pad that needs to be properly soldered too.

You can do that with low temperature solder paste and a heat gun. Just solder the middle pad like that and then do the pins by hand using a sort of drag soldering

@NickAlexeev alien abduction issues

hm, cmu200s from the netherlands on ebay...

Hello all. I am a freshman year college student currently looking to major in Electrical Engineering. What courses should I do in the summer for a better CV?

you should think about a career in carpentry

@PlasmaHH What a warm reception

@W5VO with enough wood that is set to fire...

of course, my smart-ass response is that device characterization isn't usually covered in freshman-level courses

"freshman" did you just assume their gender

@Asmyldof Thanks for the advice. I'll get a heat gun.

@PlasmaHH Unless they come with a full test and option report, they're not mine

@Allenph If you have low-temp solder (cheap Bismuth based) you could even reflow in a kitchen hot-air oven at 150 degC

@Asmyldof Meh. $35 for some Yee-Haa Chinese reflow station. I use their $18 Hakko 936 clone with a Hakko tip and it works just fine. Besides...I already pissed my girlfriend off making etching acid in the house. I doubt she'll take kindly to me putting stuff in the oven.

Did TS turn a new leaf?

@EwokNightmares He's managed not to get in many shouting matches lately, but his answers are still often off topic and self-congratulatory.

Do you think his trolling is unintentional or just sort of a natural side effect of his eccentric personality?

It says he's top 0.26% this year

@EwokNightmares He does manage to get some upvotes.

@EwokNightmares Only because the people of the sort made me largely quit main

:-P

It's been so long since I remember him being a controversy now. It's weird how many years have gone by on this thing.

This feels like a moment to say "Poopin' Time" and be silent for 10 minutes, even if it really isn't and it usually doesn't take that long.... soooo.... Poopin' Time!

Did he ever come in this chat?

(I deny all allegations of actually going out to buy milk)

Yup he did

Not that long ago last

I see a lot of people just using plain old solder and the rework station. Do I really need the special solder paste or just a ton of flux?

@Allenph A drop of solder molten on a hot PCB and letting a chip fall on neatly with a tiny bit of flux works reasonably well too

Well shit. I guess I just bought a superfluous rework station.

But a safer and easier way is a small dot of one of those $3 Chinese pastes under the thermal pad and heating it all up until the flux bubbles out and the chip seats itself neatly

@Allenph No, because in either case you want the PCB to be hot while you also heat the chip to get the best effect

In fact the method that is easier without a rework station is the paste

@Asmyldof What do you mean? How would you see the flux boil off if you...you know...can't see the flux boil off?

@Allenph It bubbles out from underneath

I'm afraid of just blasting that thing with hot air...

They're like $10 a pop. :p

But you'll also see the chip seat itself flatter against the PCB, then you angle it straight on the pads, let it cool off and if it doesn't fall off, the pad is fine

The PLL chip isn't getting hot, it just needs ground for low noise

So the quality of the connection is only required up to "stays in place and conducts at low resistance"

I figured it was just a ground plane for noise reduction. I'm confused though...from the videos I've seen and stuff I've read it seems all methods include blasting the chip itself with hot ait.

So if you don't want to heat the chip directly, because you think (possibly rightly) that your Chinesy hot air thing will overshoot too much, you probably really want solder paste, as then you can (not the best way, but CAN) only heat the PCB and wait for the chip to seat neatly and then wait another 20 seconds before turning the heat off

What do you think a reflow oven does?

Cuddle the chip while it heats the PCB only to 295 degrees?

or 275 or 285 or whatever the profile it's set to

For the solder to adhere properly both metal surfaces need to be a bit above the melting temperature

The best way to make that happen is heating both objects, if you heat only the PCB with too little power, you may even make the chip last VERY long at 190+ degrees, which is much more likely to kill something containing only semiconductor and wire-bonds than a short peak a bit too much above 300 degrees.

The whole idea of the sharp reflow peak is quickly make everything on the outside fckn hot, while the energy doesn't penetrate to the chip before everything is cooled down again

If you dunk your hand very shortly in 95°C water and immediately in cold water 10 minutes, it'll likely heal on its own. If you hold your hand under 70°C for an hour, it's very unlikely your hand will ever function the same way again.

I see what you mean. I was just told when I was younger (not sure where) that really I should even be clipping a heat-sink on transistor leads when I was soldering...it scares me to risk an expensive chip.

I see how just heading the PCB could result in the equivalent of a cold solder joint. I was also wondering about that.

And don't go and take my simile literally and drop hot PCBs in water, or you'll break every single component on it.

Just convection, natural or very mildly forced

Wow, I must seem really retarded.

I have seen graduated trade schoolers and university engineers come up with that little gem of an idea

Nothing to do with being retarded, just a sense of materials, or not.

What do you think about doing it the opposite way? Putting a small bed of solder on the center ground plane, then placing the chip and heating?

Been dawdling along for 20 minutes now. I was going to buy milk in my pretend poop break

Bye.

Then drag soldering. (Or even giving each pin the same treatment as the ground plane...then it will probably snap into place because of surface tension.)

Thanks.

Much better to just do the middle pad first, so you can try with a little pressure when it's cold if you have a cold joint or not

All the others you can see

Until you have a decent reflow system, it's better to make sure about things you might otherwise assume.

Bye now

Bye. Thanks again.

@PlasmaHH Or somebody is writing a novel about abduction issues.

Is there a place you can buy busted ICs to practice with, or is it better just to find the cheapest QFN I can go d and ruin 20 of them?

broken ICs don't do much to pracitce with...

I should have clarified. Soldering practice. Specifically that QFN practice I talked about with Asmyldof above.

it isn't that hard that you have to practice it a lot. desolder something from broken stuff, solder it back on, and then do it with real parts.

btw, the diode test mode of this multimeter is next to useless :/

Maybe I'm overly worried about it I guess.

yes you are

Not trying to overwrite other advice but is the QFN soldering for prototype? I would focus on heating the board so hot you can't touch it with the solder melting on the pads and then when it is heated up, quickly hold the IC with tweezers and shortly preheat the bottom of its pins with the heat gun for 5-10 seconds and then quickly but stably drop it with the gun not on it momentarily (so it don't blow away) and then press it down in place with tweezers and then blow on it longer

@EwokNightmares I've read about people doing it that way, but if you do it the way Asmyldof was suggesting, exactly placement shouldn't be a problem; it should lock itself in more or less correctly.

Thanks though.

Just sharing my experiences from back in the day before I made my money with making 3 day prototypes and trial runs, whether or not you follow that is up to you and who knows, maybe I was doing it less than optimal all those years.

Goodnight now.

Not quite slept all that well today